About

Assistant Professor of Practice
Department of Computer Science
The George Washington University

Human-Centered AI for EEG/BCI, Clinical EEG, and Multimodal Sensor Data

I develop AI/ML methods for clinical EEG, brain-computer interaction, and multimodal sensor data, with a focus on robust, interpretable learning for real-world human-centered decision-support settings. My work connects machine learning, human-centered AI, and health-facing AI applications, while addressing core technical questions in temporal modeling, uncertainty, and model behavior.

Research

My research uses EEG and related sensor data as demanding testbeds for AI and machine learning. I study how models can capture temporal structure, remain reliable under noisy and changing conditions, and support human interpretation rather than functioning as opaque predictors. Across this work, I combine methodological development with application areas in clinical EEG, EEG-based brain-computer interaction, and multimodal sensor analysis.

My current work spans original method development, health-facing AI, and student-centered collaborative research. A central thread is building interpretable and robust models for temporally structured human data, especially where model reliability matters for scientific interpretation, accessibility, and decision support.

My group works with undergraduate and master’s students on human-centered AI, EEG/BCI, clinical EEG, and multimodal sensor-data projects.

Research Themes

Clinical EEG and Health-Facing AI

I study machine learning methods for clinically meaningful EEG analysis, including work that uses clustering and representation learning to identify patterns linked to functional and clinical outcomes. This area anchors my broader interest in health-facing AI and decision-support settings.

EEG-Based Brain-Computer Interaction

I develop and evaluate AI methods for EEG-based brain-computer interaction, including classification, control, and human-in-the-loop systems. This work helps surface broader AI questions about robustness, user variation, and interpretation in interactive settings.

Multimodal Sensor and Time-Series Learning

I use EEG, wearable, interaction, and related sensor data to study temporal modeling, uncertainty, and real-world variation. This theme connects methodological questions in machine learning to broader human-centered data problems.

Human-Centered and Interpretable AI

I am interested in AI systems that help people reason with data, not just generate predictions. My work emphasizes model behavior, uncertainty, transparency, and the relationship between technical performance and human use.

Selected Publications

  1. Identifying clinically and functionally distinct groups among healthy controls and first episode psychosis patients by clustering on EEG patterns
    Frontiers in Psychiatry, 2020. First-author clinical EEG machine learning work developed through NIMH-supported research participation. [PDF]

  2. Multi-class time continuity voting for EEG classification
    International Conference on Brain Function Assessment in Learning, 2020. Best Paper Award; a method for improving EEG classification by modeling temporal continuity. [PDF]

  3. GViT: Combining Convolutional and Transformer Layers for Spatial-Temporal EEG Analysis
    HCI International, 2025. A student-coauthored paper supporting current work in spatial-temporal EEG modeling. [PDF]

  4. Using EEG to distinguish between writing and typing for the same cognitive task
    International Conference on Brain Function Assessment in Learning, 2020. An original EEG study connecting machine learning with human behavior and interaction. [PDF]

  5. Advancing EEG-based gaze prediction using depthwise separable convolution and enhanced pre-processing
    HCI International, 2024. Student-coauthored work on EEG-based gaze prediction and model improvement. [PDF]

  6. EEG4Home: A human-in-the-loop machine learning model for EEG-based BCI
    HCI International, 2022. A human-in-the-loop BCI paper reflecting my interest in interactive, interpretable AI systems. [PDF]

  7. Generative AI tools in higher education: A meta-analysis of cognitive impact
    CHI Extended Abstracts, 2025. A highly cited 2025 CHI extended abstract that broadens the homepage view of my AI research beyond EEG and sensor-data settings. [PDF] [Google Scholar]

Additional Publications

Core AI, Clinical EEG, and EEG/BCI

Human-Centered, Applied, and Systems

Review, Survey, and Student-Centered Research

Students and Mentoring

I mentor undergraduate and master’s students through project-based AI/ML research, with an emphasis on reproducible baselines, careful evaluation, and student-led publication pathways.

  • 40+ undergraduate and master’s students mentored

  • 10+ student-coauthored papers

  • 20+ capstone teams supervised

  • 5 student awards or fellowships

My teaching and research are closely linked: students often move from coursework into mentored research, reproducible baselines, collaborative writing, and applied AI projects.

Teaching

My teaching emphasizes a pipeline from core concepts to projects to research. I use structured labs, transparent rubrics, staged milestones, and project-based work to help students build strong technical foundations and then apply them to open-ended AI and data problems.

News

  • 2026: New HCI International papers on sleep EEG, accessible EEG game control, deployable neural directional control, and transformer-based EEG analysis.

  • 2025: Student-coauthored SIGKDD paper on EEG-controlled directional games and custom-developed BCI game evaluation.

  • 2025: CHI Extended Abstract on generative AI tools in higher education.

  • 2024: Student-coauthored SIGKDD paper on generative AI models for undergraduate research summarization.

  • 2023-2026: Program Board Member and Session Chair, HCI International.

Service

  • Program Board Member and Session Chair, HCI International (HCII)

  • 100+ peer reviews across AI, HCI, EEG/BCI, and related conference and journal venues

Service work supports my broader commitment to AI, HCI, and interdisciplinary computing communities.

Contact